Detecting gas leaks can be a challenge, particularly in large buildings or industrial sites where numerous locations could be the leak sources. As such, researchers from TU Delft, the University of Barcelona and Harvard University have developed a swarm of tiny autonomous drones that can detect and localize gas leak sources in indoor environments.
Currently, human firefighters are tasked with identifying leaks using gas sensing instruments, which can take a considerable amount of time, exposing firefighters to considerable risk. Likewise, gas sensors are less accurate than animal noses in detecting small amounts of gas and staying sensitive to quick changes in gas concentration.
"We are convinced that swarms of tiny drones are a promising avenue for autonomous gas source localization," said Guido de Croon, professor at the Micro Air Vehicle laboratory of TU Delft. "The drones' tiny size makes them very safe to any humans and property still in the building, while their flying capability will allow them to eventually search for the source in three dimensions. Moreover, their small size allows them to fly in narrow indoor areas. Finally, having a swarm of these drones allows them to localize a gas source quicker, while escaping local maxima of gas concentration in order to find the true source."
However, these properties make it hard to instill the drones with necessary artificial intelligence (AI) for autonomous gas source localization as the onboard sensing and processing is limited. Researchers found that operating drones in a swarm brings its own challenges since the drones must be aware of each other to avoid collision and collaborate.
"Actually, in nature there are ample examples of successful navigation and odor source localization within strict resource constraints," said Bart Duisterhof, a doctorate student at TU Delft. "Just think of how fruit flies with their tiny brains of ~100,000 neurons infallibly locate the bananas in your kitchen in the summer. They do this by elegantly combining simple behaviors such as flying upwind or orthogonally to the wind depending on whether they sense the odor. Although we could not directly copy these behaviors due to the absence of airflow sensors on our robots, we have instilled our robots with similarly simple behaviors to tackle the task."
How they did it
Researchers developed a new algorithm patterned after insects. As long as no drone has sensed any gas, the drones spread out as much as possible over the environment, while avoiding obstacles and each other. If one of the drones senses gas at a location, it communicates with the others and they collaborate with each other to find the gas source.
The algorithm, called particle swarm optimization (PSO), was originally modeled after the social behavior and motion of bird flocks. Each drone moves based on its own perceived highest gas concentration location and an inertia in its current moving direction. The PSO has the advantage to require measures of gas concentration and it allows the swarm to ignore local maxima that may occur in environments.
"This research shows that swarms of tiny drones can perform very complex tasks," Guido said. "We hope that this work forms an inspiration for other robotics researchers to rethink the type of AI that is necessary for autonomous flight."